SECOND LAW ANALYSIS OF SOLID OXIDE FUEL CELLS A THESIS SUBMITTED TO THE GRADUATE SCHOOL OF NATURAL AND APPLIED SCIENCES OF THE MIDDLE EAST TECHNICAL UNIVERSITY BY BAŞAR BULUT IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE IN THE DEPARTMENT OF MECHANICAL ENGINEERING SEPTEMBER 2003
130
Embed
SECOND LAW ANALYSIS OF SOLID OXIDE FUEL CELLS A ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
SECOND LAW ANALYSIS OF SOLID OXIDE FUEL CELLS
A THESIS SUBMITTED TO
THE GRADUATE SCHOOL OF NATURAL AND APPLIED SCIENCES
OF
THE MIDDLE EAST TECHNICAL UNIVERSITY
BY
BAŞAR BULUT
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF
MASTER OF SCIENCE
IN
THE DEPARTMENT OF MECHANICAL ENGINEERING
SEPTEMBER 2003
iii
ABSTRACT
SECOND LAW ANALYSIS OF SOLID OXIDE FUEL CELLS
Bulut, Başar
M.S., Department of Mechanical Engineering
Supervisor : Assoc. Prof. Dr. Cemil Yamalı
Co-Supervisor : Prof. Dr. Hafit Yüncü
September 2003, 111 pages
In this thesis, fuel cell systems are analysed thermodynamically and electrochemically.
Thermodynamic relations are applied in order to determine the change of first law and
second law efficiencies of the cells, and using the electrochemical relations, the
irreversibilities occuring inside the cell are investigated.
Following this general analysis, two simple solid oxide fuel cell systems are examined.
The first system consists of a solid oxide unit cell with external reformer. The second
law efficiency calculations for the unit cell are carried out at 1273 K and 1073 K, 1 atm
and 5 atm, and by assuming different conversion ratios for methane, hydrogen, and
oxygen in order to investigate the effects of temperature, pressure and conversion ratios
on the second law efficiency. The irreversibilities inside the cell are also calculated and
iv
graphed in order to examine their effects on the actual cell voltage and power density of
the cell.
Following the analysis of a solid oxide unit cell, a simple fuel cell system is modeled.
Exergy balance is applied at every node and component of the system. First law and
second law efficiencies, and exergy loss of the system are calculated.
Keywords: Exergy, Solid oxide fuel cell, Second law efficiency
v
ÖZ
KATI OKSİT YAKIT HÜCRELERİNİN İKİNCİ KANUN ANALİZİ
Bulut, Başar
Yüksek Lisans, Makina Mühendisliği Bölümü
Tez Yöneticisi : Doç. Dr. Cemil Yamalı
Ortak Tez Yöneticisi : Prof. Dr. Hafit Yüncü
Eylül 2003, 111 sayfa
Bu tez çalışmasında, yakıt hücresi sistemleri termodinamiksel ve elektrokimyasal olarak
analiz edilmiştir. Hücrelerin birinci ve ikinci kanun verimlerinin değişiminin
incelenmesi için genel termodinamik bağıntılar kullanılmış, elektrokimyasal bağıntılar
kullanaran hücre içinde meydana gelen tersinmezlikler incelenmiştir.
Bu genel analizin ardından, iki basit katı oksit yakıt hücresi sistemi incelenmiştir. Birinci
sistem dış düzenleyicili bir katı oksit hücresinden oluşmaktadır. İkinci kanun verim
hesaplamaları, sıcaklık, basınç ve değişme oranlarının ikinci kanun verimine etkilerini
inceleyebilmek amacıyla 1273 K ve 1073 K, 1 atm ve 5 atm, ve de metan, hidrojen ve
oksijen için değişik değişme oranları varsayılarak tamamlanmıştır. Hücre içerisindeki
vi
tersinmezlikler de hesaplanmış ve fiili hücre voltajına ve hücrenin güç yoğunluğuna olan
etkilerini inceleyebilmek için grafikleri çizilmiştir.
Katı oksit hücre analizini, basit bir katı oksit sisteminin modellenmesi takip etmiştir.
Ekserji denklemi, sistem içindeki her parça ve düğüm noktasına uygulanmıştır. Birinci
ve ikinci kanun verimleri ve sistemin ekserji kayıpları hesaplanmıştır.
Anahtar Kelimeler: Ekserji, Katı oksit yakıt hücresi, İkinci kanun verimi
vii
To My Parents
viii
ACKNOWLEDGMENTS
I would like to thank to my supervisor Assoc. Prof. Dr. Cemil Yamalı for his guidance
and insight throughout the research. I express sincere appretiation to my co-supervisor
Prof. Dr. Hafit Yüncü for his guidance, suggestions and comments.
I express sincere thanks to my family for their support and faith in me, and for their
understanding in every step of my education.
ix
TABLE OF CONTENTS
ABSTRACT ........................................................................................................... iii
ÖZ ........................................................................................................................... v
ACKNOWLEDGMENTS....................................................................................... viii
TABLE OF CONTENTS ........................................................................................ ix
LIST OF TABLES .................................................................................................. xiii
LIST OF FIGURES................................................................................................. xvi
LIST OF SYMBOLS .............................................................................................. xix
1.1 Classification of fuel cells ................................................................... 9
1.2 Application fields of fuel cells............................................................. 10
4.1 Properties of SOFC materials ............................................................... 55
5.1 Mixture compositions for ideal case ( i.e. 14=CHη , 1
2=Hη ,
12=Oη ) ……………………………………………………………… 84
5.2 Mixture compositions for 9.04=CHη , 9.0
2=Hη , 9.0
2=Oη ……….. 84
5.3 Mixture compositions for 9.04=CHη , 9.0
2=Hη , 8.0
2=Oη ……….. 84
5.4 Mixture compositions for 9.04=CHη , 9.0
2=Hη , 7.0
2=Oη ……….. 85
5.5 Mixture compositions for 9.04=CHη , 8.0
2=Hη , 8.0
2=Oη ……… 85
5.6 Mixture compositions for 9.04=CHη , 8.0
2=Hη , 7.0
2=Oη ……….. 85
5.7 Mixture compositions for 9.04=CHη , 7.0
2=Hη , 7.0
2=Oη ………... 86
5.8 Calculated second law efficiencies as functions of conversion ratios ( T = 1273 K, P = 1 atm, I = 8000 A / m² ) ……………………. 87
5.9 Calculated second law efficiencies as functions of conversion
ratios ( T = 1073 K, P = 1 atm, I = 8000 A / m² )…………………….87
xiii
xiv
5.10 Calculated second law efficiencies as functions of conversion
ratios ( T = 1273 K, P = 1 atm, I = 6000 A / m² ) …………………. 88
5.11 Calculated second law efficiencies as functions of conversion ratios ( T = 1073 K, P = 1 atm, I = 6000 A / m² ) …………………. 88
5.12 Calculated second law efficiencies as functions of conversion ratios ( T = 1273 K, P = 1 atm, I = 4000 A / m² ) …………………. 89
5.13 Calculated second law efficiencies as functions of conversion ratios ( T = 1273 K, P = 1 atm, I = 4000 A / m² ) …………………. 89
5.14 Calculated second law efficiencies as functions of conversion ratios ( T = 1273 K, P = 1 atm, I = 2000 A / m² ) …………………. 90
5.15 Calculated second law efficiencies as functions of conversion ratios ( T = 1073 K, P = 1 atm, I = 2000 A / m² ) …………………. 90
5.16 Calculated second law efficiencies as functions of conversion ratios ( T = 1273 K, P = 5 atm, I = 8000 A / m² ) …………………. 91
5.17 Calculated second law efficiencies as functions of conversion ratios ( T = 1073 K, P = 5 atm, I = 8000 A / m² ) …………………. 91
5.18 Calculated second law efficiencies as functions of conversion ratios ( T = 1273 K, P = 5 atm, I = 6000 A / m² ) …………………. 92
5.19 Calculated second law efficiencies as functions of conversion ratios ( T = 1073 K, P = 5 atm, I = 6000 A / m² ) …………………. 92
5.20 Calculated second law efficiencies as functions of conversion ratios ( T = 1273 K, P = 5 atm, I = 4000 A / m² ) …………………. 93
5.21 Calculated second law efficiencies as functions of conversion ratios ( T = 1073 K, P = 5 atm, I = 4000 A / m² ) …………………. 93
5.22 Calculated second law efficiencies as functions of conversion ratios ( T = 1273 K, P = 5 atm, I = 2000 A / m² ) …………………. 94
5.23 Calculated second law efficiencies as functions of conversion ratios ( T = 1073 K, P = 5 atm, I = 2000 A / m² ) …………………. 94
5.24 Heat exchanger design conditions and results ……………………...102
5.25 Calculated molar chemical compositions of gas streams at each node……………………………………………………………103
xv
5.26 Net work output of the SOFC operating at Ta = 1000 K, Tc = 900 K, I = 1000 A/m² …………………………………..……104
5.27 The first and second law efficiencies of model 2 …………………...104
5.28 The comparison of input, output, and loss of energy and exergy in the system. Energy and exergy values are normalized relative to the lower heating value and chemical exergy of the fuel, respectively ……………………………………………………106
xvi
LIST OF FIGURES
FIGURES
1.1 Schematic of an individual fuel cell................................................. ..... 2
1.2 Direct energy conversion with fuel cells in comparison to
to indirect energy conversion ............................................................... 3
1.3 Fuel cell power plant processes .......................................................... 5
1.4 Stacking of individual fuel cells ......................................................... 6
2.1 An open thermodynamic system with single inlet and outlet ............. 24
2.3 Schematic of the system used to calculate the second law efficiency of the simple fuel cell ………………………………….... 30
2.4 Comparison between fuel cell first law and second law efficiency changes with temperature and Carnot efficiency change with temperature……………………………………………… 32
3.1 Voltage change with current density for a simple fuel cell operating at about 40°C, and at standard pressure ................................ 35
3.2 Voltage change with current density for a solid oxide fuel cell operating at about 800°C ................................................................ 36
3.3 The film thickness theory .................................................................... 43
3.4 Types of diffusion through the pores: ( a ) Knudsen diffusion, ( b ) molecular diffusion, ( c ) transition diffusion................................ 47
xvii
4.1 Schematic of simulation model 1 ........................................................ 58
4.2 Schematic of simulation model 2 ........................................................ 59
5.1 Activation polarization change with current density ( T = 1273 K )…………………………………………………………. 73
5.2 Ohmic polarization change with current density ( T = 1273 K ) …… 73
5.3 Concentration polarization change with current density ( T = 1273 K )…………………………………………………………. 74
5.4 Change in the cell voltage and the power density with current density ( T = 1273 K )………………………………………… 74
5.5 Calculated polarization effects with current density ( T = 1273 K )…………………………………………………………. 75
5.6 Calculated power density, cell voltage and polarizations with current density ( 1273 K )…..…………………………………… 76
5.7 Activation polarization change with current density ( T = 1073 K ) ………………………………………………………. 77
5.8 Ohmic polarization change with current density ( T = 1073 K ) ..….... 78
5.9 Concentration polarization change with current density ( T = 1073 K )…………………………………………………………. 78
5.10 Change in the cell voltage and the power density with current density ( T = 1073 K )……………………………………… 79
5.11 Calculated polarization effects with current density ( T = 1073 K )………………………………………………………. 80
5.12 Calculated power density, cell voltage and polarizations with current density ( 1073 K )……………………………………… 81
5.13 Second law efficiency with current density at P = 1 atm, and P = 5 atm, conversion ratios are 100% and 90% respectively, T = 1273 K …………………………………………………………. 95
5.14 Second law efficiency with current density at P = 1 atm, and P = 5 atm, conversion ratios are 100% and 90% respectively, T = 1073 K …………………………………………………………. 96
xviii
5.15 Second law efficiency with current density at T = 1273 K, and T = 1073 K, conversion ratios are 100% and 90% respectively, P = 1 atm……………………………………………………………. 96
5.16 Second law efficiency with current density at T = 1273 K, and T = 1073 K, conversion ratios are 100% and 90% respectively, P = 5 atm……………………………………………………………. 97
5.17 Heat requirement of the components of the SOFC system with methane reforming rate ( Tr,i = 1100 K )…………………………… 99
5.18 Heat release of the combustion processes in the afterburner With respect to the mole number of the fuel that is burnt………….. 99
5.19 Comparison of heat release by methane and hydrogen with reformer efficiency …………………………………………………100
5.20 Change in fuel utilization rate with reformer efficiency …………...101
5.21 System 2 operating at 1 atm, 90% reformer efficiency, 75% fuel utilization rate, fuel inlet temperature is 1000 K, air inlet temperature is 900 ………………………………………………......105
xix
LIST OF SYMBOLS
a Activity C Concentration ( mole / m³ ) D Diffusion coefficient ( m² / s ) e Energy of molecular interaction ( ergs ) E Cell voltage F Faraday constant ( = 96487 kJ / V.kmole electrons ) G Gibbs free energy ( jJ ) i Current density ( A / m² ) io Exchange current density ( A / m² ) I Current density ( A / m² ) IL Limiting current density ( A / m² ) İ Irreversibility ( kJ / s ) J Mass flux ( kg / s ) K Equilibrium constant n Number of moles ne Electrons transferred per reaction M Molecular mass p Partial pressure ( atm ) P Pressure ( atm ) Q Heat ( kJ ) R Universal gas constant ( = 8.3145 kJ / kmole K ) Re Area specific resistance ( Ω / m² ) T Temperature ( K ) w Thickness ( µm ) W Work ( kJ ) We Electrical work ( kJ ) X Mole fraction
xx
Greek Letters α Transfer coefficient δ Thickness of the diffusion layer ( m ) ε Porosity η Polarization ( V ) ηI First law efficiency ηII Second law efficiency µ Chemical potential ( kJ / mole ) ρ Resistivity ( Ω cm ) ξ Tortuosity σ Collision diameter ( Ả ) ΩD Collision integral nased on the Lennard-Jones potential Subscripts a Anode A A specie B B specie c Cathode k Knudsen diffusion P Products R Reactants rev Reversible (eff) Effective Superscripts I Inlet condition
1
CHAPTER 1
INTRODUCTION
1.1 Definition of a Fuel Cell
A fuel cell is an electrochemical device which can continuously convert the free energy
of the reactants ( i.e. the fuel and the oxidant ), which are stored outside the cell itself,
directly to electrical energy. The basic physical structure or building block of a fuel cell
consists of an electrolyte layer in contact with two porous electrodes; the anode or the
fuel electrode, where the fuel that feeds the cell is oxidised, and the cathode or the
oxygen ( or air ) electrode, where the reduction of molecular oxygen occurs, on either
side. A schematic representation of a fuel cell with the reactant / product gases and the
ion conduction flow directions through the cell is shown in Figure 1.1 [1].
In a fuel cell, gaseous fuels are fed continuously to the anode ( negative electrode ) and
an oxidant ( i.e. oxygen from air ) is fed continuously to the cathode ( positive electrode
). The electrochemical reactions take place at the electrodes, where an electric current is
produced. The ion species and its transport direction can differ. The ion can either be
negative or positive, which means that the ion carries either a negative or a positive
charge.
Figure 1.1: Schematic of an individual fuel cell [1].
The basic principles of a fuel cell are similar to the electrochemical batteries, which are
involved in many activities of dailylife. The main difference between the batteries and
the fuel cells is that, in the case of batteries, the chemical energy is stored in substances
located inside them. When this energy has been converted to electrical energy, the
battery must be thrown away ( primary batteries ) or recharged appropriately ( secondary
batteries ). In a fuel cell, on the other hand, since the chemical energy is provided by a
fuel and an oxidant stored outside the cell in which the chemical reaction takes place, the
electrical energy is produced for as long as the fuel and oxidant are supplied to the
electrodes. Figure 1.2 shows the comparison of direct energy conversion with fuel cells
to indirect conversion.
2
Chemical energy of the fuel(s)
Electrical energy conversion
Thermal and/or mechanical energy conversion
Figure 1.2: Direct energy conversion with fuel cells in comparison to indirect energy conversion.
Gaseous hydrogen has become the fuel of choice for most applications, because of
its high reactivity when suitable catalysts are used, its ability to be produced from
hydrocarbons and its high energy density when stored cryogenically for closed
environment applications. Similarly, the most common oxidant is gaseous oxygen,
which is readily and economically available from air, and easily stored in a closed
environment.
The electrolyte not only transports dissolved reactants to the electrode, but also
conducts ionic charge between the electrodes and thereby completes the cell electric
circuit. It also provides a pysical barrier to prevent the mixing of the fuel and oxidant gas
streams.
The porous electrodes in the fuel cells provide a surface site where gas/liquid
ionization or de-ionization reaction can take place. In order to increase the rates of
3reactions, the electrode material should be catalytic as well as conductive, porous rather
4
.2 Fuel Cell Plant Description
he fuel and oxygen from the air are combined
.3 Fuel Cell Stacking
r to produce the required voltage level. The
lar plate between
conditions, and an excellent electronic conductor. [2]
than solid. The catalytic function of electrodes is more important in lower temperature
fuel cells and less so in high temperature fuel cells, because ionization reaction rates are
directly proportional with temperature ( i.e. increase with temperature ). The porous
electrodes also provide a physical barrier that separates the bulk gas phase and the
electrolyte.
1
In the fuel cell, hydrogen produced from t
to produce dc power, water, and heat. In cases where CO and CH4 are reacted in the cell
to produce H2, CO2 is also a product. These reactions must be carried out at a suitable
temperature and pressure for fuel cell operation. A system must be built around the fuel
cells to supply air and clean fuel, convert the power to a more usable form
( i.e. AC power ), and remove the depleted reactants and heat that are produced by the
reactions in the cells. Figure 1.3 shows a simple rendition of a fuel cell power plant.
Beginning with fuel processing, a conventional fuel ( natural gas, other gaseous
hydrocarbons, methanol, coal, etc. ) is cleaned, then converted into a gas containing
hydrogen. Energy conversion occurs when dc electricity is generated by means of
individual fuel cells combined in stacks or bundles. A varying number of cells or stacks
can be matched to a particular power application. Finally, power conditioning converts
the electric power from DC into regulated DC or AC for consumer use [1].
1
Individual fuel cells are combined in orde
schematic of stacking of individual fuel cells is given in Figure 1.4. [2]
Anode – electrolyte – cathode sections are connected in series by a bipo
the cathode of the cell and the anode of the other cell. The bipolar plate must be
impervious to the fuel and oxidant gases, chemically stable under reducing and oxidizing
Fuel Processor
Power Section
Power Conditioner
Natural Gas
H Gas
5
igure 1.4 is a representation of a flat plate cell. Tubular solid oxide cells are stacked in
different way. There may be other arrangements for stacking as well, provided that the
Figure 1.3: Fuel Cell Power Plant Processes
F
a
interconnectors are impervious to the gases and are excellent electronic conductors, as
explained.
2-richDC Power
AC Power
ir
Exhaust Gas
A
Usable Heat
Steam
Figure 1.4: Stacking of individual fuel cells. [2]
.4 Characteristics of Fuel Cells
These advantages of fuel cells are grouped
.4.1 Efficiency
ical energy is converted directly into electrical energy. Since direct energy
1
Fuel cells offer advantages in many fields.
into categories and are briefly explained in this section.
1
Chem
conversion doesn’t require a preliminary conversion into heat, this conversion is not
subject to the limitations of Carnot cycle, and thermal efficiencies of as high as 90% are
theoretically possible. This direct energy conversion from chemical energy to electrical
6
7
.4.2 Flexibility in Power Plant Design
e o low voltage level of an individual cell, it is
.4.3 Manufacturing and Maintenance
a low as engines. The whole system of the fuel
.4.4 Noise
l s no moving parts. It runs quietly, does not vibrate, does not generate
energy does not require any mechanical conversion, such as boiler-to-turbine and
turbine-to-generator systems. The efficiency of a cell is not dependent upon the size of
the cell. A small cell operates with an efficiency equivalent to a larger one, consequently
can be just as efficient as large ones. This is very important in the case of the small local
power generating systems needed for combined heat and power systems.
1
In ord r to obtain a desired voltage, due t
necessary to connect a number of cells in series. The current delivered by an individual
cell is proportional to the geometrical area of the electrode. The electrode may be
increased in size, or alternatively, several cells may be connected in parallel to increase
the current. These cell groups may also be connected in series or parallel to yield high
currents at high voltages. The cells need not be localized in one place, thus providing
flexibility in weight distribution and space utilization. This characteristics is most
convenient from a design viewpoint.
1
The m nufacturing cost of fuel cells is as
cells can be manufactured by mass production methods. There are no moving parts in a
cell, hence sealing problems are minimum and no bearing problems exist. Because of
these, fuel cells require little or no maintenance. Corrosion, on the other hand, is a
serious problem, especially for high temperature cells.
1
A fue cell ha
gaseous pollutants [2]. This characteristics is very important in military and
communication applications.
8
.4.5 Heat
e inefficiencies in a fuel cell may manifest themselves as heat. With proper
.4.6 Low Emissions
y main fuel cell reaction, when hydrogen is the fuel, is water,
ffer can be listed as
el flexibility
ty
pability
General negative features of fuel cells and fuel cell plants can be listed as follows :
iliar technology to the power industry
1
The el ctrical
design of a cell, efficiency can be maximized and heat can be minimized [3].
1
The b -product of the
instead of carbon dioxide, nitrogen oxides, sulfur oxides, and particulate matter inherent
to fossil fuel combustion, which means a fuel cell can be essentially a “zero emission”
device. This is fuel cells’ main advantage when used in vehicles, as there is a
requirement to reduce vehicle emissions, and even eliminate them within cities.
However, it should be noted that, at present, emissions of CO2 are nearly always
involved in the production of the hydrogen needed as the fuel [4].
Some other characteristics that fuel cells and fuel cell plants o
follows :
• Fu
• Cleanliness
• Size flexibili
• Cogeneration ca
• Site flexibility
• Cost
• Unfam
9
1.5 Types of Fuel Cells and Their Fields of Applications
Fuel cells are usually classified, according to their operating temperatures, into low,
medium, and high temperature fuel cells. Table 1.1 gives an overview of the fuel cell
technologies presently under development [5].
Table 1.1: Classification of fuel cells. [5]
Fuel Cell Type
Operating
Temperature
[oC]
Fuel Oxidation
Media
Typical Unit
Sizes
[kWe]
Alkaline 70 - 100 H2 Oxygen << 100
Protone Exchange
Membrane 50 - 100
H2 and reformed
H2
Oxygen
from air 0,1 - 500
Phosphoric Acid 160 - 210 H2 reformed from
natural gas
Oxygen
from air
5 - 200
(plants up to
5,000)
Molten Carbonate 650
H2 and CO from
internal reforming
of natural/coal gas
Oxygen
from air
800 - 2,000
(plants up to
100,000)
Solid Oxide 800 - 1000
H2 and CO from
internal reforming
of natural/coal gas
Oxygen
from air 2.5 - 100,000
10
The most probable application fields for the different types of fuel cells are presented in
Table 1.2.
Table 1.2: Application fields of fuel cells.
Fuel Cell Type Fields of Application
Alkaline Space applications and special military applications
Protone Exchange
Membrane
Stationary applications for direct hydrogen use
Stationary applications for power and heat production
Mobile applications for buses, service vehicles
Mobile applications for railroad systems
Mobile applications for passenger cars
Phosphoric Acid Stationary applications for power and heat production
Mobile applications for railroad systems
Molten Carbonate Stationary applications for combined power and vapor production
Stationary applications for utility use
Solid Oxide
Stationary applications for power and heat production
Stationary applications for utility use
Mobile applications for railroad systems
CHAPTER 2
THERMODYNAMICS OF FUEL CELLS
2.1 Some Fundamental Relations
2.1.1 TdS Equations and Maxwell Relations
Consider a simple compressible system undergoing an internally reversible process. An
energy balance for this simple compressible system, in the absence of overall system
motion and gravity effect, can be written in differential form as follows;
..int..int revrev WdUQ δδ += ( 2.1 )
The only mode of energy transfer by work that can occur as a simple compressible
system undergoes quasiequilibrium processes is associated with volume change and is
given by ∫ [6]. Therefore, the work is given by pdV
pdVW rev =..intδ ( 2.2 )
The equation for entropy change on a differential basis is given by
..int revTQdS δ
= ( 2.3 )
11
By rearrangement,
TdSQ rev =..intδ ( 2.4 )
Substituting Eqs.2.2 and 2.4 into Eq.2.1 and rearranging the terms gives the first TdS
equation;
pdVTdSdU −= ( 2.5 )
Enthalpy is, by definition,
pVUH += ( 2.6 )
On a differential basis,
VdppdVdUpVddUdH ++=+= )( ( 2.7 )
Rearranging the terms results,
VdpdHpdVdU −=+ ( 2.8 )
Substituting Eq.2.8 into Eq.2.5 and rearranging the terms gives the second TdS equation;
VdpTdSdH += ( 2.9 )
The TdS equations on a unit mass basis can be written as
pdvTdsdu −= (2.10)
vdpTdsdh += (2.11)
or on a per mole basis as
vpdsTdud −= (2.12)
pvdsTdhd += (2.13)
From these two fundamental relations, two additional equations may be formed by
defining two other properties of matter.
The Helmholtz function ψ is defined by the equation
Tsu −=Ψ (2.14)
12
Forming the differential dψ results,
sdTTdsduTsddud −−=−=Ψ )( (2.15)
Substituting Eq.2.10 into Eq.2.15 gives
sdTpdVd −−=Ψ (2.16)
The Gibbs function is defined by the equation
Tshg −= (2.17)
Forming the differential dg results,
sdTTdsdhTsddhdg −−=−= )( (2.18)
Substituting Eq. 2.1 into Eq. 2.8 gives
sdTvdpdg −= (2.19)
From the comparison of Eq.2.16 and Eq.2.19, one can conclude that Gibbs function
carries out reactions at constant pressure and temperature, while Helmholtz function
does at constant volume and temperature. Since it is more practical to carry out reactions
at constant pressure and temperature, Gibbs function is more useful and is preferred in
calculations.
As a result, the summary of these four important relationships among properties of
simple compressible systems are collected and presented below:
pdvTdsdu −= (2.10)
vdpTdsdh += (2.11)
sdTpdVd −−=Ψ (2.16)
sdTvdpdg −= (2.19)
These equations are referred to as TdS ( or Gibbsian ) equations. Note that the variables
on the right-hand sides of these equations include only T, s, p, and v.
13
Consider three thermodynamic variables represented by x, y, and z. Their functional
relationship may be expressed in the form x = x ( y, z ). The total differential of the
dependent variable x is given by the equation
dzzxdy
yxdx
yz
⎟⎠⎞
⎜⎝⎛∂∂
+⎟⎟⎠
⎞⎜⎜⎝
⎛∂∂
= (2.20a)
If in Eq.2.20a we denote the coefficient of dy by M and the coefficient of dz by N,
Eq.2.20a becomes
NdzMdydx += (2.20b)
Partial differentiation of M and N with resprect to z and y, respectively, leads to
zyx
zM
y ∂∂∂
=⎟⎠⎞
∂∂ 2
(2.21a)
and
yzx
yN
z ∂∂∂
=⎟⎟⎠
⎞∂∂ 2
(2.21b)
If these partial derivatives exist, it is known from the calculus that the order of
differentiation is immaterial, so that
zy yN
zM
⎟⎟⎠
⎞∂∂
=⎟⎠⎞
∂∂ (2.21c)
When Eq.2.21c is satisfied for any function x, then dx is an exact differential. Eq.2.21c
is known as the test for exactness. [7]
Since only properties are involved, each TdS equation is an exact differential exhibiting
the general form of Eq.2.20a. Underlying these exact differentials are functions of the
form u ( s, v ), h ( s, p ), ψ ( v, T ), and g ( T, p ), respectively.
The differential of the function u ( s, v ) is
dvvuds
sudu
sv⎟⎠⎞
⎜⎝⎛∂∂
+⎟⎠⎞
⎜⎝⎛∂∂
= (2.22)
14
Comparing Eq.2.22 to Eq.2.10 results,
vsuT ⎟⎠⎞
⎜⎝⎛∂∂
= (2.23a)
svup ⎟⎠⎞
⎜⎝⎛∂∂
=− (2.23b)
The differential of the function h ( s, p ) is
dpphds
shdh
sp⎟⎟⎠
⎞⎜⎜⎝
⎛∂∂
+⎟⎠⎞
⎜⎝⎛∂∂
= (2.24)
Comparing Eq.2.24 to Eq.2.11 results,
pshT ⎟⎠⎞
⎜⎝⎛∂∂
= (2.25a)
sphv ⎟⎟⎠
⎞⎜⎜⎝
⎛∂∂
= (2.25b)
The differential of the function ψ ( v, T ) is
dTT
dvv
dvT⎟⎠⎞
⎜⎝⎛∂Ψ
+⎟⎠⎞
⎜⎝⎛∂Ψ∂
=Ψα (2.26)
Comparing Eq.2.26 to Eq.2.16 results,
Tvp ⎟
⎠⎞
⎜⎝⎛∂Ψ∂
=− (2.27a)
vTs ⎟
⎠⎞
⎜⎝⎛∂Ψ∂
=− (2.27b)
The differential of the function g ( T, p ) is
dTTgdp
pgdg
pT
⎟⎠⎞
⎜⎝⎛∂∂
+⎟⎟⎠
⎞⎜⎜⎝
⎛∂∂
= (2.28)
15
Comparing Eq.2.28 to Eq.2.19 results,
Tpgv ⎟⎟⎠
⎞⎜⎜⎝
⎛∂∂
= (2.29a)
pTgs ⎟⎠⎞
⎜⎝⎛∂∂
=− (2.29b)
Since each of the four differentials is exact and similar to Eq.2.20a, referring to
Eq.2.21c, the following relations can be written:
vs sp
vT
⎟⎠⎞
⎜⎝⎛∂∂
−=⎟⎠⎞
⎜⎝⎛∂∂
(2.30)
ps sv
pT
⎟⎠⎞
⎜⎝⎛∂∂
=⎟⎟⎠
⎞⎜⎜⎝
⎛∂∂
(2.31)
Tv vs
Tp
⎟⎠⎞
⎜⎝⎛∂∂
=⎟⎠⎞
⎜⎝⎛∂∂
(2.32)
Tp ps
Tv
⎟⎟⎠
⎞⎜⎜⎝
⎛∂∂
−=⎟⎠⎞
⎜⎝⎛∂∂
(2.33)
This set of equations is referred to as the Maxwell relations.
2.1.2 Partial Molal Properties
In general, the change in any extensive thermodynamic property X of a multicomponent
system can be expressed as a function of two independent intensive properties and size
of the system. Selecting temperature and pressure as the independent properties and the
number of moles n as the measure of size, this change in any extensive thermodynamic
property X can be expressed as follows:
inpTi inTnp
dnnXdp
pXdT
TXdX
j,,,,∑ ⎟⎟
⎠
⎞⎜⎜⎝
⎛∂∂
+⎟⎟⎠
⎞⎜⎜⎝
⎛∂∂
+⎟⎠⎞
⎜⎝⎛∂∂
= (2.34)
16
where the subscript nj denotes that all n’s except ni are held fixed during differentiation.
The last term on the right-hand side of the Eq.2.34 is defined as the partial molal
property iX of the ith component in a mixture. Therefore, the partial molal property
iX is, by definition
jnpTii n
XX,,
⎟⎟⎠
⎞∂∂
= (2.35)
The extensive thermodynamic property X, can be expressed in terms of the partial molal
property iX as
∑=j
ii XnX1
(2.36)
Eq.2.36 can be referred in order to evaluate the change in volume on mixing of pure
components which are at the same temperature and pressure. Selecting V as the
extensive property X in Eq.2.36 the total volume of the pure components before mixing
is
∑=
=j
iiicomp vnV
1,0. (2.37)
where iv ,0 is the molar specific volume of pure component i. The volume of the mixture,
using Eq.2.36, is
∑=
=j
iiimix vnV
1. (2.38)
where iv is the partial molal volume of component i in the mixture. Hence, the volume
change on mixing is given by
∑∑==
−=−=∆j
iii
j
iiicompmixmixing vnvnVVV
1,0
1.. (2.39a)
or
17
( )∑=
−=∆j
iiiimixing vvnV
1,0 (2.39b)
Selecting U, H, and S as the extensive properties, the similar results can be obtained as
follows:
( )∑=
−=∆j
iiiimixing uunU
1,0 (2.40a)
( )∑=
−=∆j
iiiimixing hhnH
1,0 (2.40b)
( )∑=
−=∆j
iiiimixing ssnS
1,0 (2.40c)
In Eqs.2.40a – c, iu ,0 , ih ,0 , and is ,0 denote molar internal energy, enthalpy, and entropy
of pure component i; iu , ih , and is denote respective partial molal properties.
2.1.3 Chemical Potential
Of the partial molal properties, the partial molal Gibbs function is particularly useful in
describing the behaviour of mixtures and solutions. This quantity plays a central role in
the criteria of both chemical and phase equilibrium. Because of its importance in study
of multicomponent systems, the partial molal Gibbs function of component i is given a
special name and symbol. It is called the chemical potential of component i and
symbolized by µi. [6]
jnpTiii n
GG,,
⎟⎟⎠
⎞∂∂
==µ (2.41)
Gibbs function can be expressed in terms of chemical potential as
∑=j
iinG1
µ (2.42)
18
The differential of G ( T, p, n1, n2, ... , nj ) can be formed as
∑ ⎟⎟⎠
⎞⎜⎜⎝
⎛∂∂
+⎟⎠⎞
⎜⎝⎛∂∂
+⎟⎟⎠
⎞⎜⎜⎝
⎛∂∂
=i
inpTinpnT
dnnGdT
TGdp
pGdG
j,,,,
(2.43)
Substituting Eqs.2.29a – b into Eq.2.43 yields,
∑=
+−=j
iii dnSdTVdpdG
1µ (2.44)
2.2 Thermodynamics of Chemical Reactions
2.2.1 Free Energy Change of Chemical Reactions
Consider the chemical reaction below;
dDcCbBaA +→+ (2.45)
The change in Gibbs function of reaction, or Gibbs free energy of the reaction, under
constant temperature and pressure, is given by the equation
BADC badcG µµµµ −−+=∆ (2.46)
where µ is the chemical potential of the species.
The maximum net work obtainable from a chemical reaction can be calculated by the
free energy change of the chemical reaction. Referring to Eq.2.17, the free energy
change of a chemical reaction is given by,
STHG ∆−∆=∆ (2.47)
2.2.2 Standard Free Energy Change of a Chemical Reaction
The chemical potential of any substance may be expressed by an equation of the form
aRTo ln+= µµ (2.48)
19
where a is the activity of the substance and µ has the value µ° when a is unity. The
standard free energy change ∆G° of the reaction 2.43 is given by
oooo BADCo badcG µµµµ −−+=∆ (2.49)
where µC° indicates the standard chemical potential of product C, and so on. Substituting
Eqs.2.48 and 2.49 into Eq.2.46 yields
bB
aA
dD
cCo
aaaa
RTGG ln+∆=∆ (2.50)
Hence, the standard free energy change of a chemical reaction is
bB
aA
dD
cCo
aaaaRTGG ln−∆=∆ (2.51)
Assuming a process at constant temperature and pressure at equilibrium, since the free
energy change for this process is zero, Eq.2.51 becomes
KRTaaaa
RTG beqB
aeqA
deqD
ceqCo lnln
,,
,, −=−=∆ (2.52)
where the suffixes eq in the activity terms indicate the values of the activities at
equilibrium, and K is the equilibrium constant for the reaction.
The importance of the knowledge of ∆G° is that it allows ∆G to be calculated for any
composition of a reaction mixture. Knowledge of ∆G indicates whether a reaction will
occur or not. If ∆G is positive, a reaction cannot occur for the assumed composition of
reactants and products. If ∆G is negative, a reaction can occur. [8]
2.2.3 Relation Between Free Energy Change in a Cell Reaction and Cell Potential
The enthalpy change of any reaction, assuming constant temperature and pressure, can
be showed as follows :
VPWQVPEH ∆+−=∆+∆=∆ (2.53)
20
If the reaction is carried out in a heat engine, then the only work done by the system
would be the expansion work,
VPW ∆= (2.54)
Hence Eq.2.51 becomes;
QH =∆ (2.55)
If the same reaction, which is under consideration is carried out electrochemically, the
only work done by the system will not be the expansion work of the gases produced, but
will also be the electrical work due to the charges being transported around the circuit
between the electrodes. The maximum electrical work that can be done by the overall
reaction carried out in a cell, where Vrev,c and Vrev,a are the reversible potentials at the
cathode and anode respectively, is given by
( )arevcreve VVneW ,,max, −= (2.56)
In the cell, n electrons are involved and the cell is assumed to be reversible
( i.e., overpotential losses are assumed to be zero ). Multiplying Eq.2.56 by the
Avogadro number, N, in order to have molar quantities gives;
reve VnFW ∆=max, (2.57)
where F is the Faraday number, and revV∆ is the difference between reversible electrode
potentials.
The only work forms assumed are the expansion work and electrical work.
VPWW e ∆+= max, (2.58)
In addition to these, assuming the process is reversible
STQ ∆= (2.59)
Substituting Eqs.2.57 – 2.59 into Eq.2.53, the enthalpy change will be
revVnFSTH ∆−∆=∆ (2.60)
21
Eq.2.60 can be rearranged as follows,
revVnFSTH ∆−=∆−∆ (2.61)
where
STHG ∆−∆=∆ (2.47)
and
revVE ∆= (2.62)
Substituting Eqs.2.47 and 2.62 into Eq.2.61 gives
nFEG −=∆ (2.63)
E, which is defined as the difference in potentials between the electrodes is called as the
electromotive force of the cell ( i.e, the reversible potential of the cell, Erev ). If both the
reactants and the products are in their standard states, Eq.2.63 can be written as,
oo nFEG −=∆ (2.64)
where E° is the standard electromotive force, or – as most commonly referred to – is the
standard reversible potential of the cell.
2.3 Nernst Equation
Let us consider the following reaction,
mMlLkK →+ (2.65)
where k moles of K react with l moles of L to produce m moles of M. Each of the
reactants and the products have an associated activity; aK , and aL being the activity of
the reactants, aM being the activity of the product. For ideal gases, activity term can be
written as
(2.66) 0p
pa =
22
p is the partial pressure of the gas, and p0 is the pressure of the cell. Eq.2.50 can be
rearranged for the the reaction given in Eq.2.65, as follows.
⎟⎟⎠
⎞⎜⎜⎝
⎛+∆=∆ l
LkK
mMo
aaaRTGG ln (2.67)
In the Eq.2.67, G∆ and oG∆ show the change in molar Gibbs free energy of formation,
and the change in standard molar Gibbs free energy of formation.
From Eq.2.63, the following relation can be written,
nFGE ∆
−= (2.68)
Substituting Eq.2.68 into Eq.2.67 gives the effect on voltage as follows,
⎟⎟⎠
⎞⎜⎜⎝
⎛−
∆−= l
LkK
mM
o
aaa
nFRT
nFGE ln (2.69)
Substituting Eq. 2.64 into Eq 2.69 yields,
⎟⎟⎠
⎞⎜⎜⎝
⎛−= l
LkK
mMo
o aaa
nFRTEE ln (2.70a)
where E° is the standard electromotive force, and Eo is defined to indicate the reversible
electric voltage. Eq.2.70a can be rewritten by substituting Eq.2.52 and Eq.2.66, as
follows.
⎟⎟⎟⎟⎟
⎠
⎞
⎜⎜⎜⎜⎜
⎝
⎛
⎟⎟⎠
⎞⎜⎜⎝
⎛⎟⎟⎠
⎞⎜⎜⎝
⎛
⎟⎟⎠
⎞⎜⎜⎝
⎛
−= l
o
L
k
o
K
m
o
M
o
pp
pp
pp
nFRTK
nFRTE lnln (2.70b)
Eq. 2.70a and 2.70b give the electromotive force in terms of product or/and reactant
activity, and is called Nernst equation. The electromotive force calculated using this
equation is known as the Nernst voltage, and is the reversible cell voltage that would
exist at a given temperature and pressure.
23
2.4 Exergy Concept
2.4.1 Exergy Balance
Exergy is defined as the maximum amount of work obtainable a substance can yield
when it is brought reversibly to equilibrium with the environment [9]. The exergy
analysis of a system is based on the second law of thermodynamics and the concept of
entropy production. In order to describe the exergy concept, a fuel cell can be modeled
as a control volume of a thermodynamic system with a single inlet and outlet, as shown
in Figure 2.1.
To
24
Figure 2.1 : An open thermodynamic system with single inlet and outlet.
inlet outlet
oQ& W&
Environment jQ&
kiTP iooo
,,1,, ,
K=
µ lj ,,1K= Tj
The goal in power producing systems is to maximize net work and efficiency. A power
plant operates according to the first and second laws of thermodynamics [10]. To
calculate the maximum work that can be produced, let us consider the system in Figure
2.1. The streams in and out of the system consist of n species with molar flow
rates , where i = 1… k. The heat transfer interactions … and properties at
the inlet and outlet are assumed to be fixed. [11] The first law for the system in Figure
2.1 can be written as;
outiini nn ,, , && 1Q& lQ&
( ) ( )∑∑∑===
−−−+−+=k
ioutioutiit
k
iiniiniit
l
jj nhhnhhWQQ
dtdE
1,,0,
1,,0,
10 &&&&& (2.71)
The second law for the same system can be written as;
( ) ( ) gen
k
ioutioutii
k
iiniinii
l
j j
j SnssnssTQ
TQ
dtdS &&&
&&+−−−+⎟
⎟⎠
⎞⎜⎜⎝
⎛+⎟⎟
⎠
⎞⎜⎜⎝
⎛= ∑∑∑
=== 1,,0
1,,0
10
0 (2.72)
In the equations above, ht,i is the total specific enthalpy and is
iit gzVhh ⎟
⎠⎞
⎜⎝⎛ ++= 2
, 21 (2.73)
½V2 is the kinetic energy, gz is the potential energy of the mass flow, and these kinetic
and potential energies may be neglected so that ht,i = hi. E is the total energy of the
system, and S is the entropy of the system. In order to make time derivatives zero,
system is assumed to be steady state, steady flow. By eliminating between Eqs.2.71
and 2.72, the exergy balance can be obtained. Therefore, the exergy balance is given by
the following equation.
0Q&
( )[ ]
( )[ ] gen
k
ioutiioutii
k
iiniiinii
l
jj
j
STnsThW
nsThQTT
&&&
&&
01
,,00
1,,00
1
01
+⋅−−+=
⋅−−+⎟⎟⎠
⎞⎜⎜⎝
⎛−
∑
∑∑
=
==
µ
µ
(2.74)
25
W& is the actual work of the system. It can be noted that entropy generation reduces the
available work, as is expected. The exergy balance for an open system in Eq.2.74 shows
that the exergies in heat flows ( the first term on the left hand side of the equation ) and
mass flows ( the second term on the left hand side of the equation ) supplied to the
system are equal to the work produced ( the first term on the right hand side of the
equation ), exergy in the outlet mass flow ( the second term on the right hand side of the
equation ), and exergy destroyed through irreversible processes ( the third term on the
right hand side of the equation ) [11]. For the steady state, steady flow system, energy
input is equal to the energy output. Due to irreversible processes, outlet exergy is always
less than the inlet exergy. Exergy destruction is a result of chemical and physical
processes that take place in the system.
Irreversibility, for a system can be described as the difference between the reversible
work ( maximum work that can be obtained ) and the actual work. Hence, from the
definition
genactrev STWWI &&&&0=−= (2.75)
The exergy balance ( Eq.2.74 ) can be used to calculate the irreversibility.
Exergy analysis requires that the environment is defined. For a general case,
environment can be assumed to be at standard temperature and pressure conditions
( i.e., T = 298 K, P = 1 atm ), but this assumption is not always the case. Environment
definition can differ from system to system.
2.4.2 Chemical Exergy
Chemical exergy is equal to the maximum amount of work obtainable when the
substance under consideration is brought from the environmental state to the dead state
by processes involving heat transfer and exchange of substances only with the
environment. [9]
Chemical exergy of a mixture is given by the following equation.
26
∑∑ +=i
iioi
oiimix xxRTxEx ln~ε (2.76)
oiε
~ is the standard chemical exergy of substance i. [9] The exergy of the mixture is
always less than the sum of the exergies of its components at the temperature and
pressure of the mixture, since the second term on the right hand side is always negative.
2.4.3 Physical Exergy
Physical exergy is equal to the maximum amount of work obtainable when the stream of
substance is brought from its initial state to the environmental state defined by Po and To
by physical processes involving only thermal interaction with the environment. [9]
Defining the environment state at Po, To, assuming the kinetic and potential energies are
negligible, the physical exergy of a substance at state P1, T1 is calculated by the
following equation.
( ) ( ooooph sThsThEx −−−= 111, ) (2.77)
2.5 Efficiency of Fuel Cells
2.5.1 Thermodynamic ( First and Second Law ) Efficiencies
In order to define the efficiency of fuel cells, let us consider a simple H2/O2 fuel cell,
operating at T = 25°C. and P = 1 atm. as shown in Figure 2.2. The inlet and outlet
conditions are assumed to be the same for simplicity.
To model the preheaters of the SOFC system ( i.e. simulation model 1 ), effectiveness-
NTU method ( ε-NTU ) is used. The counter flow type heat exchangers are selected as
the preheaters. The heat transfer coefficients are set equal to 0.05 kW/m²K. The areas of
the heat exchangers must be determined before the calculation steps, in order to use the
effectiveness-NTU method. Because of this reason, the preheaters are first sized
according to the assumed operating conditions. After sizing, the performance analysis
can be carried out.
The outlet temperatures of the preheaters depend on the inlet conditions. To define the
effectiveness of the heat exchanger, the maximum possible heat transfer rate, qmax must
be determined. This maximum possible heat transfer rate depends on the minimum heat
capacity rate. [19] To determine the minimum heat capacity rate, the heat capacity rates
of the cold and hot gas streams are calculated from the following equations.
∑= cpcc cmC ,& (4.19a)
∑= hphh cmC ,& (4.19b)
The subscripts “c” and “h” indicate cold and hot gas streams respectively. Comparing
the heat capacity rates of the cold and hot gas streams, one with the lower value is
determined as the minimum heat capacity rate, while the other is determined as the
maximum heat capacity rate. Following this, the heat capacity rates ratio is given as,
63
max
min
CCCr = (4.20)
Hence, the maximum possible heat transfer rate can be calculated by,
( )icih TTCq ,,minmax −= (4.21)
Th,i is the inlet temperature of the hot gas stream, and Tc,i is the inlet temperature of the
cold gas stream.
The number of transfer units ( NTU ) is a dimensionless parameter that is used for heat
exchanger analysis and is defined as,
minCUANTU = (4.22)
U is the heat transfer coefficient; A is the area of the heat exchanger.
The effectiveness of the counter flow type heat exchangers can be calculated from the
following equation.
( )[ ]([ )]rr
r
CNTUCCNTU−−−
−−−=
1exp11exp1
ε (4.23)
The effectiveness, ε, is defined as the ratio of the actual heat transfer rate to the
maximum possible heat transfer rate. Therefore, with known effectiveness and
maximum possible heat transfer rate, the actual heat transfer rate can be calculated.
maxqq ⋅= ε (4.24)
The energy balance for the cold and hot gas streams can be written as follows.
( )icocc TTCq ,, −= (4.25a)
( )ohihh TTCq ,, −= (4.25b)
Tc,o and Th,o represents the outlet temperature of the cold gas stream and the outlet
temperature of the hot gas stream respectively. As a result, the cold and hot gas streams’
outlet temperatures can be found by rearranging Eqs.4.25a – b, as follows.
64
cicoc C
qTT += ,, (4.26a)
hihoh C
qTT −= ,, (4.26b)
The hot gas stream inlet temperature is also the SOFC outlet temperature. Because of
this, the hot gas inlet temperatures of both of the preheaters are affected by the SOFC
outlet temperature. For this reason, the calculation steps for the heat exchanger model
are repeated until the outlet temperatures of the gas streams from the preheaters
converge.
4.7 Calculation Procedure
In the calculations, two different models are simulated as mentioned. The reason for this
is first of all to investigate the effects on the fuel cell unit and the changes of effects with
different conditions. After this study, using a simple model, a general thermodynamic
analysis for every component of the fuel cell system is made.
The model 1 is simulated in order to examine
• The effect of reformer efficiency and fuel utilization rate on the fuel cell second
law efficiency,
• The effect of pressure on the fuel cell second law efficiency,
• The comparison of the results obtained with different temperatures,
• The polarization effects on fuel cell voltage,
• The effect of temperature on fuel cell voltage and power density.
On the other hand, the more complex system, model 2, is simulated in order to make a
full thermodynamic analysis on each component of a simple SOFC system. The first
model deals with the SOFC unit with external reformer only. But the second model
examines the use of the waste heat at the exit of the SOFC unit to heat the reformer and
65
66
vaporizer for the reactions to take place, and to heat the fuel and air inlet nodes to the
required temperatures. For these purposes afterburner and two preheaters are added to
the system. The preheaters are designed using the heat exchanger model.
4.7.1 General Assumptions
Before the calculation steps, the general assumptions and conditions for the simulation
models are listed as follows;
• Steady flow throughout the nodes,
• Kinetic and potential energy changes are neglected,
• Frictional effects are assumed to be negligible,
• Fuel supplying rate is 1 kg/s,
• Operating temperature is constant throughout the fuel cell unit ( i.e. fuel cell unit
inlet, exit and reformer inlet temperatures are all the same ) ( for model 1 only ),
• Afterburner is assumed to be isothermal, i.e. the inlet and outlet temperatures are
the same ( for model 2 only ),
• The reactions in the afterburner are assumed to be complete ( for model 2 only),
• The environment is at standard temperature and pressure conditions, i.e. 298 K,
and 1 atm.
The calculations are carried out under the given assumptions. Since simulation model 1
will investigate the effects of the reformer efficiency, fuel utilization rate, the operating
pressure on the fuel cell efficiency and the effects of polarizations on fuel cell voltage
and power density with changing current density, there is no need to make additional
assumptions for these variables for model 1. Hence, the general assumptions are applied
to the simulation model 1 directly. For simulation model 2, the following variables will
be assumed to be known initially, and the values will be given for these variables.
• The operating pressure of the fuel cell system,
• Percentage of theoretical air,
• Reformer efficiency,
• Fuel utilization rate,
• Fuel cell fuel inlet and air inlet temperatures.
4.7.2 Calculation Steps for Simulation Model 1
Before thermodynamic analysis of the fuel cell unit, using the electrochemical model,
the voltage drops in fuel cell unit due to polarizations, their effect on fuel cell voltage
and power density are calculated.
The model is first assumed to be at T = 1273 K , and then at T = 1073 K. The
electrochemical model to calculate the polarizations is used for both temperature values.
By this way, the effect of temperature on fuel cell power density, voltage and
polarizations can be considered. The anode and cathode exchange current densities are
taken as Io,a = 70000 A / m² , Io,c = 24000 A / m² for T = 1273 K, and Io,a = 5000 A / m²,
Io,c = 2000 A / m² for T = 1073 K [12].
After the calculation of the polarization effects using the electrochemical model,
thermodynamic analysis of the fuel cell unit is made.
Exergy efficiency is defined as the ratio of the work output to the maximum work output
of the system. Hence, it can be written as follows,
max
max
WWW lost
II−
=η (4.27)
The maximum work available for the system is given by the Eq.2.45. Writing Eq.2.45 in
terms of molar properties,
STHGW o∆−∆=∆=max (4.28)
The work losses are due to incomplete reactions in the reformer and fuel cell unit, and
polarizations occurring in the fuel cell unit. Since hydrogen is reformed from methane
67
gas, the moles of hydrogen depend on reformed methane, and so do the moles of
oxygen. The reactions occurring in reformer and at the anode and cathode of the fuel cell
are written assuming methane is completely reformed and the fuel cell reactions are
ideal.
( )reformerOHCOHOHCH 22224 2.042.2 ++→+ (4.29a)
( )anodeeOHOH −− +→+ 8444 22
2 (4.29b)
( )cathodeOeO −− →+ 22 482 (4.29c)
The overall reaction of the system is the sum of reactions 4.17a – c.
2224 22 COOHOCH +→+ (4.30)
The maximum work of the system is calculated using complete conversion ratios
( i.e. 14=CHη , 1
2=Hη , and 1
2=Oη ).§ In reality, maximum values of the conversion
ratios will not be unity, but smaller. These differences cause incomplete reactions and
hence work loss.
The reformer reaction, since it is an irreversible combustion process, is a complete loss
of work. Hence work loss can be calculated using the Gibbs function.
STHGW refloss ∆−∆=∆=, (2.45)
The work loss for the incomplete reactions can be calculated the same way, since the
conversion ratio decreases the available work output of the reaction.
STHGW incloss ∆−∆=∆=., (2.45)
In the electrolyte cracks may occur. This means that the cell is short circuited, some fuel
reacts irreversible with oxygen and this produces hot spots. [20] An assumption of 1 %
of the hydrogen react in this way, the work loss due to cracks is given by the following
equation.
68
§ 224
,, OHCH ηηη are the conversion ratios of CH4, H2, and O2, respectively. This value indicates how much of CH4 is reformed and how much of H2, and O2 is reacted in the reaction. Since maximum work is the reversible work, the ideal conversion ratios are assumed, which is not the case realistically.
reactionHreactionHcrackloss STHGW −− ∆−∆=∆⋅=22
)(01.001.0, (4.31)
The polarizations occurring in the fuel cell causes work loss. From the electrochemical
model, the polarizations are calculated with current density. Hence, operating fuel cell
with an assumed current density, the work loss due to polarizations can be calculated.
The work loss caused by polarizations is,
69
)( OhmConcActirrloss FnW ηηη ++⋅⋅=, (4.32)
The number of electrons, n, is 8 in ideal case. But, that depends on the conversion ratios
of methane, hydrogen, and oxygen, and is calculated according to the given conversion
ratios.
As seen from Figure 4.1, there are four nodes in model 1. At each node of the system,
there is a mixture of gases, and since partial pressures apply, the composition of the
mixtures is to be figured out. For this reason, before analyzing the system
thermodynamically, the mixture compositions at each node of the system regarding to
the conversion ratios of each reaction are determined. With the mixture components
known at each node, the related properties of the gases can be calculated.
After thermodynamic analysis is accomplished, the heat release in the fuel cell will also
be examined.
4.7.3 Calculation Steps for Simulation Model 2
The complete reforming of methane in the reformer and complete reaction of hydrogen
in the fuel cell stack are not realistic.
Reforming of methane is an endothermic reaction and requires heat input to the system
for the reaction to take place. The vaporizing of liquid water to water vapor is also an
endothermic reaction and requires heat input. Because of these heat requirements of the
reformer and vaporizer reactions, there must be some fuel left at the exit of the fuel cell.
The afterburner is added to the exit of the fuel cell stack in order to burn the remaining
fuel and release heat required for the reformer and vaporizer reactions.
70
Before calculations, the heat requirements of the components are investigated.
Regarding to these results, the conversion ratios for the reformer and the fuel cell unit
reactions are determined.
Before stepping into calculations, using the heat exchanger model, the heat areas of the
two preheaters are calculated. The system is modeled stoichiometrically, that is the air
supplied to the system is 100% theoretical ( i.e. there is no extent of air ).
With the calculations explained above, the reformer efficiency due to the conversion
ratio of methane, the fuel utilization rate due to the conversion ratio of hydrogen are
determined, and the heat transfer area needed in the two preheaters are calculated.
Modeling the inlet conditions at standard temperature and pressure ( i.e. T = 298 K,
P = 1 atm ), under the general assumptions and the assumed and calculated values for the
reformer efficiency, fuel utilization rate and heat transfer area, the simulation model 2
can be analyzed. The calculation steps for this model are numbered and listed in order as
follows:
1. Since the methane and hydrogen conversion ratios are determined, and there is a
complete combustion in the afterburner, the required air amount by the SOFC
stack and the afterburner to sustain the system operation is calculated.
2. The molar chemical compositions of the flow streams at each node of the system
are determined.
3. Using the known molar chemical compositions of the flow streams at each node
and the fuel supplying rate, the mass flow rates at each node are determined.
4. The temperature values of the fuel and air inlet must be assumed. With known
fuel and air inlet temperatures, the electrochemical model can be used in order to
determine the electrical work output, the exit temperature, and the heat rejection
of the SOFC stack. The first law and second law efficiencies of the SOFC stack
are also calculated.
5. Heat exchanger model is used to determine the exit temperatures of the heat
exchangers. The reformer inlet temperature is calculated at this step. The cold
gas stream outlet temperature of the preheater 2, which is also the assumed air
71
inlet temperature of the SOFC, is compared with the calculated result. Since the
exit temperatures of the preheaters depend on the exit temperature of the SOFC
stack, the iteration is continued until the outlet temperature of the preheater 2
converges.
6. Since temperatures and molar chemical compositions at each node are known,
applying the energy and exergy balance at each component, the enthalpy and
exergy at each node are calculated.
The exergy balance for a component is derived in Eq.2.74 and is used to calculate the
exergy destruction in the component. The exergy of a mixture at a node, where the
temperature and molar chemical compositions are known, is calculated by using
Eqs.2.76 and 2.77.
72
CHAPTER 5
RESULTS AND DISCUSSION
5.1 Results of Simulation Model 1
5.1.1 Electrochemical Model Analysis
The first electrochemical results for the simulation model 1 are calculated at T = 1273 K.
The changes in activation, ohmic, and concentration polarizations with current density
are given in Figures 5.1, 5.2, and 5.3, respectively. The change in cell voltage and power
density due to the polarizations is given in Figure 5.4. The change in all kinds of
polarizations with current density is grouped and shown in a single graph in Figure 5.5.
The calculated power density, cell voltage, and polarizations versus current density is
given in Figure 5.6.
0
0,01
0,02
0,03
0,04
0,05
0 2000 4000 6000 8000 10000 12000
Current Dens ity ( A / m ² )
Act
ivat
ion
Pola
rizat
ion
( V )
Anode Cathode
T = 1273 K
Figure 5.1: Activation polarization change with current density ( T = 1273 K ).
0
0,01
0,02
0,03
0,04
0,05
0,06
0 2000 4000 6000 8000 10000 12000
Current Density ( A / m ² )
Ohm
ic P
olar
izat
ion
( V )
T = 1273 K
Figure 5.2: Ohmic polarization change with current density ( T = 1273 K ).
73
0
0,05
0,1
0,15
0,2
0,25
0,3
0 2000 4000 6000 8000 10000 12000Current Density ( A / m² )
Con
cent
ratio
n Po
lariz
atio
n ( V
)
Anode Cathode
T = 1273 K
Figure 5.3: Concentration polarization change with current density ( T = 1273 K ).
0
0,2
0,4
0,6
0,8
1
1,2
0 2000 4000 6000 8000 10000 12000
Current Density ( A / m ² )
Cel
l Vol
tage
( V
)
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
Pow
er D
ensi
ty (
W /
m² )
Cell Voltage Pow er Density
T = 1273 K
Figure 5.4: Change in the cell voltage and the power density with current density ( T = 1273 K ).
74
0
0,01
0,02
0,03
0,04
0,05
0,06
0,07
0,08
0,090,
1
020
0040
0060
0080
0010
000
1200
0
Cur
rent
Den
sity
( A
/ m² )
Voltage ( V )
Anod
e Ac
tivat
ion
Pola
rizat
ion
Cat
hode
Act
ivatio
n Po
lariz
atio
nO
hmic
Pol
ariza
tion
Anod
e C
once
ntra
tion
Pola
rizat
ion
Cat
hode
Con
cent
ratio
n Po
lariz
atio
n
T =
1273
K
Figu
re 5
.5: C
alcu
late
d po
lariz
atio
n ef
fect
s with
cur
rent
den
sity
( T
= 12
73 K
).
75
0
0,2
0,4
0,6
0,81
1,2
020
0040
0060
0080
0010
000
1200
0Cu
rren
t Den
sity
( A
/ m² )
Cell Voltage ( V )
01000
2000
3000
4000
5000
6000
7000
8000
9000
1000
0
Power Density ( W / m² )
Ano
de A
ctiv
atio
n Po
lariz
atio
nCa
thod
e A
ctiv
atio
n Po
lariz
atio
nO
hmic
Pol
ariz
atio
n
Ano
de C
once
ntra
tion
Pola
rizat
ion
Cath
ode
Conc
entra
tion
Pola
rizat
ion
Cell V
olta
ge
Pow
er D
ensi
ty
T =
1273
K
Figu
re 5
.6: C
alcu
late
d po
wer
den
sity
, cel
l vol
tage
, and
pol
ariz
atio
ns w
ith c
urre
nt d
ensi
ty
( T =
127
3 K
).
76
For a second condition, temperature is decreased to T = 1073 K. The same results
graphed for T = 1073 this time.
The changes in activation, ohmic, and concentration polarizations with current density
are given in Figures 5.7, 5.8, and 5.9. The change in cell voltage and the power density
due to the polarizations is given in Figure 5.10. The change in all kinds of polarizations
with current density is grouped and shown in a single graph in Figure 5.11. The
calculated power density, cell voltage, and polarizations versus current density is shown
in Figure 5.12.
0
0,05
0,1
0,15
0,2
0,25
0,3
0,35
0 2000 4000 6000 8000 10000 12000
Current Dens ity ( A / m ² )
Act
ivat
ion
Pola
rizat
ion
( V )
Anode Cathode
T = 1073
Figure 5.7: Activation polarization change with current density ( T = 1073 K ).
77
Ohmic
0
0,04
0,08
0,12
0,16
0,2
0 2000 4000 6000 8000 10000 12000Current Dens ity ( A / m ² )
Ohm
ic P
olar
izat
ion
( V )
T = 1073 K
Figure 5.8: Ohmic polarization change with current density ( T = 1073 K ).
0
0,05
0,1
0,15
0,2
0,25
0,3
0 2000 4000 6000 8000 10000 12000 14000
Current Dens ity ( A / m ² )
Con
cent
ratio
n Po
lariz
atio
n ( V
)
Anode Cathode
T = 1073 K
Figure 5.9: Concentration polarization change with current density ( T = 1073 K ).
78
0
0,2
0,4
0,6
0,8
1
1,2
0 2000 4000 6000 8000 10000 12000
Current Dens ity ( A / m ² )
Cel
l Vol
tage
( V
)
0
500
1000
1500
2000
2500
3000
3500
4000
4500
Pow
er D
ensi
ty (
W /
m² )
Cell Voltage Pow er Density
T = 1073 K
Figure 5.10: Change in the cell voltage and the power density with current density ( T = 1073 K ).
79
0
0,1
0,2
0,3
0,4
0,5
020
0040
0060
0080
0010
000
1200
014
000
Cur
rent
Den
sity
( A
/ m² )
Voltage ( V )
Ano
de A
ctiv
atio
n P
olar
izat
ion
Cat
hode
Act
ivat
ion
Pol
ariz
atio
nO
hmic
Pol
ariz
atio
nA
node
Con
cent
ratio
n P
olar
izat
ion
Cat
hode
Con
cent
ratio
n P
olar
izat
ion
T =
1073
K
Figu
re 5
.11:
Cal
cula
ted
pola
rizat
ion
effe
cts w
ith c
urre
nt d
ensi
ty (
T =
1073
K ).
80
0
0,2
0,4
0,6
0,81
1,2
020
0040
0060
0080
0010
000
1200
014
000
Cur
rent
Den
sity
( A
/ m² )
Voltage ( V )
0500
1000
1500
2000
2500
3000
3500
4000
4500
Power Density ( W / m² )
Ano
de A
ctiv
atio
n P
olar
izat
ion
Cat
hode
Act
ivat
ion
Pol
ariz
atio
nO
hmic
Pol
ariz
atio
nA
node
Con
cent
ratio
n P
olar
izat
ion
Cat
hode
Con
cent
ratio
n P
olar
izat
ion
Cel
l Vol
tage
Pow
er D
ensi
ty
T =
1073
K
Figu
re 5
.12:
Cal
cula
ted
pow
er d
ensi
ty, c
ell v
olta
ge a
nd p
olar
izat
ions
with
cur
rent
den
sity
( T
= 10
73 K
).
81
82
The cathode activation polarization is obviously higher than the anode activation
polarization. This is because of the cathode’s lower exchange current density. Both the
cathode and anode activation polarizations increase almost linearly at T = 1273 K.
Ohmic polarization changes linearly with increasing current, since it is related to the
material thickness. Electrolyte section has the main effect on ohmic polarization due to
its material’s high resistivity. The effect of ohmic polarization can be changed by
changing the thickness of materials or by using another material instead.
Anode concentration polarization is higher than the cathode concentration polarization at
low current densities. But after a critical region, cathode concentration polarization
increases exponentially and goes to infinity as can be seen from the graphs. Regarding to
Figures 5.3 and 5.9, for the cell working at T = 1273 K, IL = 10788 A / m², and for the
one working at T = 1073 K, IL = 11900 A / m² are the limiting current densities.
Figures 5.4 and 5.10 give the change of the cell voltage and the power densities with
current density. The lower the temperature, the more the cell voltage drops, and the less
the power density is. This result is just as expected. In chapter 3, it was explained that
the efficiency of fuel cells decreases with increasing temperature, but on the other hand,
as it can be observed from the results, as the temperature increases, the drop in cell
voltage decreases, and power density increases.
Figures 5.5 and 5.11 show the polarization effects in a single graph for both
temperatures. The results show that at lower temperatures, the cathode activation
polarization becomes significant and has the most effect on voltage drop. The ohmic and
anode activation polarizations behave similarly at lower temperatures. The anode
concentration polarization, on the other hand, becomes significant at high temperature
values, while cathode concentration polarization has very little effect at low
temperatures and low current densities. At high temperature, cathode concentration
polarization increases exponentially. Anode and cathode activation polarizations depend
on temperature much more than the other types of polarizations. The increase in
83
activation polarization with decreasing temperature is the highest in all types of
polarizations.
Figures 5.6 and 5.12 give a better understanding of the effect of temperature on the fuel
cell voltage and the power density. For the cell operating at T = 1273 K, at I ≈ 10400 A /
m², the power density reaches its peak point P ≈ 8889 W / m². For the cell operating at T
= 1073 K, at I ≈ 8000 A / m², the power density reaches its peak point P ≈ 3955 W / m².
At I = 6000 A / m², for the cell working at T = 1273, cell voltage is, E = 0.989 V, while
for the cell working at T = 1073 K, cell voltage is, E = 0.621 V. There is a significant
difference between the two cell voltages.
As a conclusion, as the fuel cell operating temperature increases, the polarizations
decrease and this results higher cell voltages and high power densities
5.1.2 Thermodynamic Analysis
The system is assumed to be at following conditions:
• T = 1273 K., P = 1 atm.
• The air consists of 20 % O2 and 80 % N2 in molar ratios.
In order to calculate the entropies at each node of the system, the partial pressures of the
molecules must be determined. For this reason, mixture compositions at each node of the
system for ideal case ( i.e. conversion ratios are unity ) is given in Table 5.1. Mixture
compositions for different conversion ratios at each node of the system are given in
Tables 5.2 – 7.
Table 5.1: Mixture compositions for ideal case ( i.e. 14=CHη , 1
The enthalpy and exergy values are calculated for each node by applying the energy
balance and the exergy balance at each component. The calculated enthalpy and exergy
values and temperatures of each node are given in Figure 5.21. The comparison of input,
output and loss of energy and exergy in the system is given in Table 5.28.
105
Figure 5.21: System 2 operating at 1 atm, 90% reformer efficiency, 75% fuel utilization rate, fuel inlet temperature is 1000 K, air inlet temperature is 900 K.
SOFC
Reactor 1831.025 kJ/s
Vaporizer 2868.29 kJ/s
Preheater 2714.113 kJ/s
Mixer
325.602kJ/s
Preheater 1 357.389 kJ/s
Reformer 993.629 kJ/s
Afterburner3529.976 kJ/s
T11=556.06 K H11=5680.777 kJ/s T7=900 K 11 Ex11=4851.519 kJ/s H7=10665.35 kJ/s
Ex7=5427.977 kJ/s T2=298 K H2= 0 Ex2= 0 2 7
Wact=32866.456 kJ/s Air Inlet Qlost=2892.613 kJ/s
Ex=2101.113 kJ/s
T10=919.79 K T6=1000 K 10 H10=16346.127 kJ/s H6=65443.051 kJ/s T4=298 K T1=298 K Ex10=10993.609 kJ/s Ex6=63238.711 kJ/s H4=50018.703 kJ/s H1=50018.703 kJ/s
Ex4=53476.663 kJ/s Ex1=52177.681 kJ/s
1 4 5
6 8
3
9
T5=1034.97 K H5=55091.112 kJ/s T9=1089.07 K T8=1089.07 K Q1=10351.939 kJ/s Ex5=56712.981 kJ/s H9=21418.536 kJ/s H8=409349.332kJ/s Ex=7519.359 kJ/s
Ex9=14587.316 kJ/s Ex8=31868.094 kJ/s
Q2=6185.36 kJ/s T3=298 K H3=0 Ex=4492.874 kJ/s Ex3=1624.584 kJ/s
Qo=2393.497 kJ/s Ex=1738.569 kJ/s
Environment, To
Water
106
Table 5.28: The comparison of input, output, and loss of energy and exergy in the system. Energy and exergy values are normalized relative to the lower heating value and chemical exergy of the fuel, respectively.
Energy Exergy
Fuel Input 100 100 Work Output 65,709 62,99 Heat to the Environment 10,568 7,359 Vaporization Process 12,366 5,497 Exhaust Gas ( Node 11 ) 11,357 9,298 Irreversibility in System Units 14,856
100 100
The results obtained show that the first law and second law efficiencies for the model
simulated are 65.71% and 62.99%, respectively. The second law efficiency of the model
is lower than the first law efficiency, because of the higher chemical exergy of the fuel
than its lower heating value.
The system is modeled with the assumption that the fuel cell is supplied 100%
theoretical air. In practice, in order to maintain a relatively high oxygen concentration at
the cathode, a high air – fuel ratio is required. This high ratio of air – fuel will increase
the irreversibility.
The exhaust gas ( node 11 ) has a high temperature. This high temperature value causes
more exergy destruction. Reducing the fuel utilization rate or methane reforming rate
can decrease this high temperature value, provided that the required heat is produced in
the afterburner to sustain the system operations. The exhaust gas of the fuel cell system
being supplied with high extent of air will have a lower temperature value, since
preheater 1 in that case will destroy more exergy.
Mixer destroyed little exergy since the inlet and outlet temperatures are the same and
irreversibility is only due to mixing of methane and water vapor.
107
The most irreversible process is the combustion in the afterburner. The system with a
high fuel cell exit temperature value would have less irreversibility in the afterburner.
The irreversibility in the SOFC stack would be decreased with a low fuel cell exit
temperature value.
Vaporization process requires heat input. This irreversibility can be eliminated by
developing a recycling fuel cell system which can be used to eliminate the need for the
vaporizer.
Methane reforming causes more exergy destruction within the system. Reducing the
combustion of fuel while reforming as much methane as possible ( hence producing as
much fuel as possible ) would increase the second law efficiency.
108
CHAPTER 6
CONCLUSION
The results obtained for the simulation models show that solid oxide fuel cells have high
second law efficiencies. As discussed before, this high efficiency values are one of the
benefits of fuel cell systems. Solid oxide fuel cell’s high operating temperature caused
less voltage drop, and therefore the irreversibilities occuring inside the fuel cell were
less.
Different configurations for the system might be studied, such as assuming different
methane – water vapor ratio, supplying some extent air, or trying different reformer
efficiency and fuel utilization rate values. Results obtained from these configurations
can be compared with the results obtained in this study to give a more general discussion
to the subject.
High operating temperature of the solid oxide fuel cell system gives the advantage to be
combined with a gas turbine system for higher efficiencies. Studies on this combined
system show that high efficiency values can be obtained this way.
The exhaust gas stream of the solid oxide fuel cell system has high exergy value. This
exhaust gas may be used for local area heating.
109
The vaporizer and afterburner components of the system destroyed more exergy when
compared with the other components of the system. Hence, different alternatives should
be tried in order to increase the second law efficiency of the system.
One of the alternatives can be the elimination of vaporizer. Since water vapor is
produced at the exit of the fuel cell stack, addition of a splitter at the outlet of the fuel
cell stack might be used to recycle the required water vapor to the reformer. The
required water vapor can be recycled from the splitter to the reformer. This arrangement
would eliminate the need for the vaporizer while increasing the system’s second law
efficiency.
Another alternative should consider the heating requirement of the reformer. Since this
heat requirement is supplied by combustion of methane and hydgoren, the fuel
utilization rate and the reformer efficiency are lower. Instead of using hydrogen, some of
the methane entering the system might directly be used to heat the reformer, hence the
reformer efficiency would be higher and system would be modeled in order to reform as
much methane as possible. This way, more methane would be reformed and more fuel
would be produced. Combustion of the fuel should therefore be reduced as much as
possible. Such a model with higher reformer efficiency and fuel utilization rate would
have high second law efficiency.
These alternatives can increase the second law efficiency, and hence offers much better
results. The future works should consider these alternative configurations.
110
REFERENCES
1. “Fuel Cell Handbook ( Fifth Edition )”, EG&G Services Parsons, Inc. Science Applications International Corporation, 2000.
2. Robert H. Perry, Don W. Green, James O. Maloney, “Perry’s Chemical Engineers’ Handbook ( Seventh Edition )”, McGraw-Hill, 1999.
3. J.R. Will Mitchell, “Fuel Cells”, Academic Press, 1963.
4. James Larminie, Andrew Dicks, “Fuel Cell Systems Explained”, John Wiley & Sons, 2000.
5. Reinhold Wurster, “PEM Fuel Cells in Stationary and Mobile Applications: infrastructural requirements, environmental benefits, efficiency advantages and economical implications”, Ludwig-Bölkow-Systemtechnik GmbH, 1997.
6. Michael J. Moran, Howard N. Shapiro, “Fundamentals of Engineering Thermodynamics ( Third Edition )”, John Wiley & Sons, 1998.
7. Jr. Kenneth Wark, “Advanced Thermodynamics For Engineers”, Mc-Graw-Hill, 1995.
8. J.O’M. Bockris, S. Srinivasan, “Fuel Cells: Their Electrochemistry”, McGraw-Hill, 1969.
9. T.J.Kotas, “The Exergy Method of Thermal Plant Analysis”, Butterworths, 1985.
111
10. Adrian Bejan, “Advanced Engineering Thermodynamics”, John Wiley & Sons, 1988.
11. Kai W. Bedringǻs, Ivar S. Ertesvǻg, Stǻle Byggstøyl, Bjørn F. Magnussen, “Exergy Analysis of Solid Oxide Fuel cell ( SOFC ) Systems”, Energy, 4, 1997, 403 – 412.
12. S.H. Chan, C.F. Low, O.L. Ding, “Energy and Exergy Analysis of Simple Solid-Oxide Fuel-Cell Power Systems”, Journal of Power Sources, 103, 2002, 188 – 200.
13. Keith R. Williams, “An Introduction to Fuel Cells”, Elsevier Pub. Co., 1966.
14. Stanley W. Angrist, “Direct Energy Conversion ( Third Edition )”, Allyn and Bacon, Boston, 1976.
15. S.H. Chan, K.A. Khor, Z.T. Xia, “A Complete Polarization Model of a Solid Oxide Fuel Cell and Its Sensitivity to the Change of Cell Component Thickness”, Journal of Power Sources, 93, 2001, 130 – 140.
16. Christie J. Geankoplis, “Mass Transport Phenomena”, Holt, Rinehart and Winston, Inc., 1972.
17. Carl Berger, “Handbook of Fuel Cell Technology”, Prentice-Hall, 1968.
18. Karl Kordesch, Günter Simader, “Fuel Cells and Their Applications”, VCH, 1996.
19. Frank P. Incropera, David P. De Witt, “Fundamentals of Heat and Mass Transfer ( Third Edition )”, John Wiley & Sons, 1990.
20. Signe Kjelstrup Ratkje, Steffen Møller-Holst, “Exergy Efficiency and Local Heat Production in Solid Oxide Fuel Cells”, Electrochimiza Acta, 38, 2/3, 1993, 447 – 453.